Skip to main content

Primary research

How many contractor websites run a chat widget

The strict count: a real chat bubble loading on the page, not just a line of copy somewhere promising live chat. If the bubble is not there when the page opens, it does not count.

20.3% of the 380 contractor websites we inspected, across roofing, remodeling and HVAC. That is 77 brands of 380.

Chat is the one capture path that works while the phone line is busy and the form feels like too much commitment. It is also the one most likely to be installed, forgotten, and left answering nobody, which is worse than not having it. Adoption is the easy number. Whether anybody is on the other end is the number nobody publishes.

The number by trade

Trade Chat widget adoption Brands Scope
Roofing 16.9% 130 strict vendor signature
Remodeling 17.9% 146 strict vendor signature
HVAC 27.9% 104 strict vendor signature

So the spread runs from HVAC at 27.9% to roofing at 16.9%. Which trade you work in changes what "normal" looks like, so read your own trade's page rather than the combined number.

Roofing

Roofing posts the lowest chat adoption of the three. Roofing demand spikes after storms, which is exactly the condition where the phone line saturates and a second channel earns its keep, so this one looks like a genuine miss rather than a defensible choice.

Full roofing breakdown for this metric

Remodeling

Remodeling sits just above roofing and well below HVAC. Remodeling has the longest sales cycle of the three, so the case for instant chat is the weakest, and the adoption rate reflects a reasonable read of the buying behaviour rather than a gap.

Full remodeling breakdown for this metric

HVAC

HVAC leads the three trades on chat adoption, and the emergency-demand pattern explains it. When the furnace dies at midnight the customer will take any channel that answers, and HVAC operators are the most likely of the three to be running an after-hours answering service that can staff it.

Full HVAC breakdown for this metric

Check your own site in ten minutes

Send a message to your own widget at nine on a Saturday night and time the reply. If the answer is Monday, the widget is collecting intent you are not converting, and an honest offline form beats a chat bubble that implies somebody is there.

What was measured
Share of brands with a vendor-detected chat widget at page load (strict).
How we measured it
We checked whether a real chat bubble loads when the page opens, rather than counting sites that merely mention live chat in their copy.
The technical version, for anyone checking our work
dom_facts vendor-signature detection at page load
Sample
380 contractor brands. Roofing 130, remodeling 146, HVAC 104.
Collection window
2026-05-10 to 2026-05-16
Combined aggregation
weighted (sum of per-trade brand counts over sum of trade sample sizes)
Who ran it
Fervor Studio, first-party capture. Nobody paid to be included or excluded.
Get My Site Inspection